The relationship between cybernetics and information is well-established, primarily through the work of Shannon. Shannon believed that his techniques could go only go as far as information, although Weaver does suggest that the insights of information transfer could eventually unlock some of the mechanisms whereby information carries deeper significance to individuals. Bateson, in defining information as 'a difference that makes a difference' effectively puts himself at one-step removed from Shannon's position, by emphasising that the important thing about information is not the stuff which is exchanged between end-points, but the relationship between those end-points as each engage in self-organising behaviour and make differences to each other in the process.
This is basically Maturana's biological position regarding information, which he sees as a 'consensual domain' of individuals self-organising, making differences that make differences to each other. It is this position which Luhmann takes as a starting point for rethinking the relationship between information, communication and meaning.
But Luhmann's other big influence, apart from Maturana, is Shannon. Like Shannon, Luhmann sees communication as a sequence of selections: selection of information to be communicated; selection of utterance; selection of meaning by the recipient. And like Shannon's theory, each selection may be seen probablistically. But Luhmann takes from Maturana the concept of organisationally-closed systems (i.e. closed systems with feedback) to argue that a separation must exist between what he terms the psychic system of the individual and the social system of communications within which individuals exist and maintain through continuing to make communications. (This is in contrast to other cybernetic theories of society like Pask's Interaction of Actors theory).
For Luhmann, what is made in the social system is 'meaning' - which is seen as the operational result of these interactions. And then Luhmann's sociological project begins by identifying different 'meaning systems' in modern society: love, education, law, economics, health, etc. Each 'meaning system' is a self-organising system oriented around a central organising principle, which is distinct in the case of each social system. In effect, this means that Luhmann had to write a lot of books! (each one about the operation of a different system of meaning).
The problem in all this is that Luhmann never identified in detail what 'meaning' really looked like. In recent years, there has been a suggestion that meaning might relate to the ability of a social system to anticipate its future states. This I find a fascinating idea (it relates in some way to the functioning of System 4 in the Viable System Model). In the mathematical work of Dubois, the simple equations of the Pearl-Verhulst equation can give us an idea of what an anticipatory system might look like, or at least how it might function in an abstract sense.
But since we live in such a data-rich time, surely we ought to be able to find some metric of anticipation between social systems within the data that we have to hand? This is my understanding of what Leydesdorff has been trying to do. Using a combination of Shannon's equations and longitudinal data from simple web-searches, the entropy of different search terms can be easily calculated. Because of the power of being able to look for occurrences in domains and co-occurrences across domains, longitudinal patterns of information transfer may be identified. It may well be that these patterns have some direct bearing on emerging meanings, and new anticipations in social systems.
What's important about this 'search for meaning' is that this data analysis reveals something which is not directly observable from looking at the surface data. But by calculating the transmission of information between domains, we can start to model the anticipatory systems that lie behind them. What is most fascinating about that is that Dubois's equations of anticipation are recursive: the fractal of the logistic map is the result. Working back from information transfer to mechanisms of anticipation that produce them could reveal fractal patterns that sit behind the data that we see.
Of course, none of this is any good unless we can use it. Rather, none of it is any good unless it results in better decision and more effective organisation. But since much of our bad decisions and poor organisation rests on a shallow interpretation of what is immediately in front of us, the opportunity to grasp the deeper structures of the data which all of our social enterprises swim in might just be worth exploring!
This is basically Maturana's biological position regarding information, which he sees as a 'consensual domain' of individuals self-organising, making differences that make differences to each other. It is this position which Luhmann takes as a starting point for rethinking the relationship between information, communication and meaning.
But Luhmann's other big influence, apart from Maturana, is Shannon. Like Shannon, Luhmann sees communication as a sequence of selections: selection of information to be communicated; selection of utterance; selection of meaning by the recipient. And like Shannon's theory, each selection may be seen probablistically. But Luhmann takes from Maturana the concept of organisationally-closed systems (i.e. closed systems with feedback) to argue that a separation must exist between what he terms the psychic system of the individual and the social system of communications within which individuals exist and maintain through continuing to make communications. (This is in contrast to other cybernetic theories of society like Pask's Interaction of Actors theory).
For Luhmann, what is made in the social system is 'meaning' - which is seen as the operational result of these interactions. And then Luhmann's sociological project begins by identifying different 'meaning systems' in modern society: love, education, law, economics, health, etc. Each 'meaning system' is a self-organising system oriented around a central organising principle, which is distinct in the case of each social system. In effect, this means that Luhmann had to write a lot of books! (each one about the operation of a different system of meaning).
The problem in all this is that Luhmann never identified in detail what 'meaning' really looked like. In recent years, there has been a suggestion that meaning might relate to the ability of a social system to anticipate its future states. This I find a fascinating idea (it relates in some way to the functioning of System 4 in the Viable System Model). In the mathematical work of Dubois, the simple equations of the Pearl-Verhulst equation can give us an idea of what an anticipatory system might look like, or at least how it might function in an abstract sense.
But since we live in such a data-rich time, surely we ought to be able to find some metric of anticipation between social systems within the data that we have to hand? This is my understanding of what Leydesdorff has been trying to do. Using a combination of Shannon's equations and longitudinal data from simple web-searches, the entropy of different search terms can be easily calculated. Because of the power of being able to look for occurrences in domains and co-occurrences across domains, longitudinal patterns of information transfer may be identified. It may well be that these patterns have some direct bearing on emerging meanings, and new anticipations in social systems.
What's important about this 'search for meaning' is that this data analysis reveals something which is not directly observable from looking at the surface data. But by calculating the transmission of information between domains, we can start to model the anticipatory systems that lie behind them. What is most fascinating about that is that Dubois's equations of anticipation are recursive: the fractal of the logistic map is the result. Working back from information transfer to mechanisms of anticipation that produce them could reveal fractal patterns that sit behind the data that we see.
Of course, none of this is any good unless we can use it. Rather, none of it is any good unless it results in better decision and more effective organisation. But since much of our bad decisions and poor organisation rests on a shallow interpretation of what is immediately in front of us, the opportunity to grasp the deeper structures of the data which all of our social enterprises swim in might just be worth exploring!
1 comment:
Post a Comment